Calibration of unified Parkinson's disease rating scale scores to Movement Disorder Society‐unified Parkinson's disease rating scale scores

Abstract

The aim of this study was to develop formulas to convert the UPDRS to Movement Disorder Society (MDS)‐UPDRS scores. The MDS‐UPDRS is a revision of the UPDRS with sound clinimetric properties. Reliable formulas to recalculate UPDRS scores into MDS‐UPDRS equivalents are pivotal to the practical transition and definitive adoption of the MDS‐UPDRS. UPDRS and MDS‐UPDRS scores were collected on 875 PD patients. A developmental sample was used to regress UPDRS scores on corresponding MDS‐UPDRS scores based on three H & Y groupings (I/II, III, and IV/V). Regression weighting factors and intercept terms provided formulas for UPDRS conversions to be tested in a validation sample. Concordance between the true MDS‐UPDRS Part scores and those derived from the formulas was compared using Bland‐Altman's plots and Lin's concordance coefficient (LCC). Significant concordance between UPDRS‐estimated MDS‐UPDRS scores was achieved for Parts II (Motor Experiences of Daily Living) (LCC = 0.93) and III (Motor Examination) (LCC = 0.97). The formulas resulted in mean differences between the true MDS‐UPDRS and estimated MDS‐UPDRS scores of less than 1 point for both Parts II and III. Concordance was not achieved for Parts I and IV (Non‐motor Experiences of Daily Living and Complications of Therapy). Formulas allow archival UPDRS Parts II and III individual patient data to be accurately transferred to MDS‐UPDRS scores. Because Part I collects data on much more extensive information than the UPDRS, and because Part IV is structured differently in the two versions, old ratings for these parts cannot be converted. © 2012 Movement Disorder Society